2025 Volume 52 Issue 3 Pages 142-149
Objective: To evaluate the probability of Barthel Index (BI) deterioration through decision curve analysis (DCA) and to examine the net benefit (NB) associated with the frequency of home-based rehabilitation (home rehab) frequency.
Methods: A regression model was developed to predict BI deterioration over 6 months based on age, sex, Charlson Comorbidity Index, Short Physical Performance Battery score, and initial BI score using data from 104 elderly patients receiving home rehab. The predictive accuracy of the model was evaluated using area under the curve (AUC). The NB of once-weekly versus twice-weekly home rehab was calculated using DCA.
Results: The predictive model achieved an AUC of 0.86. DCA showed that for patients with a BI deterioration probability of ≥61% and ≥81%, twice-weekly home rehab increased NB by 31.1% and 48.0%, respectively.
Conclusion: Patients with a higher probability of BI deterioration may derive greater NB from receiving multiple weekly home rehab sessions.
